Groups > comp.programming.threads > #3891

ThreadpoolX that scales very well is here..

Path	csiph.com!eternal-september.org!feeder.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From	Intelli2 <intelli2@mama.com>
Newsgroups	comp.programming.threads
Subject	ThreadpoolX that scales very well is here..
Date	Fri, 13 Oct 2017 18:39:08 -0400
Organization	A noiseless patient Spider
Lines	60
Message-ID	<orr18c$lfp$2@dont-email.me> (permalink)
Mime-Version	1.0
Content-Type	text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding	7bit
Injection-Date	Fri, 13 Oct 2017 18:39:08 -0000 (UTC)
Injection-Info	reader02.eternal-september.org; posting-host="f1428d66b964250dc5044036e8578199"; logging-data="22009"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX184W7OjC3L2L5bp7Cj77m6DUqpVD7qj/1c="
User-Agent	Mozilla/5.0 (Windows NT 10.0; WOW64; rv:52.0) Gecko/20100101 Thunderbird/52.4.0
Content-Language	en-US
X-Mozilla-News-Host	news://news.eternal-september.org:119
Cancel-Lock	sha1:YiMWVwMgOJ/nxaWDcBELpUkfwc4=
Xref	csiph.com comp.programming.threads:3891

Show key headers only | View raw

Hello,

I have ported the scalable counting network algorithm to Delphi and
FreePascal, and i have used it inside my following ThreadpoolX engine,
and now it scales very well, here is the papers of the scalable counting
network algorithm:

http://people.csail.mit.edu/shanir/publications/AHS.pdf

and also read the following (the counting network is truly scalable,
please look at the graph inside the paper):

http://people.csail.mit.edu/shanir/publications/HLS.pdf

you have to look at my following Threadpool engine that scales very well
now, because it is very powerful now, here is what it supports:

- I have used scalable counting networks to make my Threadpool engine
scale very well.

- The worker threads enter in a wait state when there is no job in the
concurrent FIFO queues - for more efficiency -

- You can distribute your jobs to the worker threads and call any method
with the threadpool's execute() method.

- It uses work-stealing to be more efficient.

- You can configure it to use stacks or FIFO queues , when you use
stacks it will be cache efficient.

- Now it can use processor groups on windows, so that it can use more
than 64 logical processors and it scales well.

- Now it distributes the jobs on multiple FIFO queues or stacks so that
it scales well.

- You can wait for the jobs to finish with the wait() method.

- It's NUMA-aware and NUMA efficient.

- And it is portable to many operating systems.

Please read the new HTML tutorial inside the zip.
You can download it from:

https://sites.google.com/site/aminer68/an-efficient-threadpool-engine-that-scales-very-well

and the one with priortities is here:

https://sites.google.com/site/aminer68/an-efficient-threadpool-engine-with-priorities-that-scales-very-well

And the Linux version is coming soon...

Thank you,
Amine Moulay Ramdane.

Back to comp.programming.threads | Previous | Next | Find similar | Unroll thread

Thread

ThreadpoolX that scales very well is here.. Intelli2 <intelli2@mama.com> - 2017-10-13 18:39 -0400

csiph-web